Trajectory privacy protection method and device based on sensitive area

Abstract

The application discloses a sensitive area-based trajectory privacy protection method and device, the method comprising: taking a preset sensitive point as a calculation point, iteratively performing a related point determination step to determine a related point set; constructing a sensitive area containing the related point set for the preset sensitive point; sequentially connecting position points in the related point set to form a sensitive sub-trajectory; adding noise to the trajectory points of the sensitive sub-trajectory by using a personalized privacy budget allocation mode of a differential privacy model; and replacing the part of the initial trajectory located in the sensitive area with the sensitive sub-trajectory after adding the noise to generate a to-be-published spatiotemporal trajectory data set. The technical scheme provided by the application realizes local accurate noise adding, fully and reasonably utilizes the privacy budget, improves the usability of data, and reduces the difficulty of data mining.

Description

Technical Field

[0001] This application relates to the field of trajectory privacy protection technology, and in particular to a trajectory privacy protection method and apparatus based on sensitive areas. Background Technology

[0002] The development of mobile internet technology and the continuous upgrading of mobile terminals with location capabilities have brought great convenience to people's lives. Location-based services such as point-of-interest (POI) lookup and recommendation, and navigation have covered all aspects of people's daily lives. At the same time, service providers collect users' trajectory data and attempt to analyze it to provide better services. However, trajectory data may contain a large amount of sensitive user information. If this sensitive information is not effectively protected against privacy breaches and is released directly, it could be vulnerable to attacks and privacy breaches.

[0003] Currently, three trajectory privacy protection methods are widely used in privacy protection: generalization, suppression, and fake trajectory methods. While these three methods protect users' trajectory data privacy to a certain extent, they neglect the utility of trajectory data after privacy protection, and excessive trajectory data protection can significantly increase the difficulty of data mining. Summary of the Invention

[0004] This application provides a trajectory privacy protection method and apparatus based on sensitive areas, which solves the technical problem in the prior art that the utility of trajectory data is ignored when protecting trajectory data privacy, leading to a significant increase in the difficulty of data mining.

[0005] In a first aspect, embodiments of this application provide a trajectory privacy protection method based on sensitive regions. The method includes: using preset sensitive points as inference points, iteratively executing a relevant point determination step to determine a relevant point set; constructing a sensitive region containing the relevant point set for the preset sensitive points; sequentially connecting the position points within the relevant point set to form a sensitive sub-trajectory; adding noise to the trajectory points of the sensitive sub-trajectory using a personalized privacy budget allocation method of a differential privacy model; replacing the portion of the initial trajectory located within the sensitive region with the noise-added sensitive sub-trajectory to generate a spatiotemporal trajectory dataset to be published; the relevant point determination step includes: adding the inference point to the relevant point set; determining whether the probability of inferring the inference point when the previous or next position point of the inference point is known is greater than the probability of directly inferring the inference point; if the determination result is yes, determining the previous or next position point of the inference point as the inference point.

[0006] In conjunction with the first aspect, in one possible implementation, constructing a sensitive region containing the relevant point set for the preset sensitive point includes: initializing the sensitive region using a minimum boundary rectangle with the latitude and longitude of the preset sensitive point and adjacent preset distance location points as boundaries; and modifying the sensitive region according to the position of each point in the relevant point set.

[0007] In conjunction with the first aspect, in one possible implementation, the differential privacy model is as follows: ;in, , The number of the initial trajectories. For the total privacy budget, A fixed-size privacy budget is obtained for each of the aforementioned sensitive sub-trajectories. For relevant points The allocated privacy budget, The privacy budget scale is defined for any location within the sensitive area. The location entropy; the privacy budget scale for any location within the sensitive area is defined as: ;in, It is the shortest Euclidean distance from the relevant point to the preset sensitive point; the position entropy is: ;in, This represents the proportion of times a user accesses a location out of the total number of times that location is accessed.

[0008] In conjunction with the first aspect, in one possible implementation, adding noise to the trajectory points of the sensitive sub-trajectory includes: determining the initial trajectory in which the sensitive sub-trajectory is located and its position within the initial trajectory; generating a first random noise and a second random noise that follow a Laplace distribution and are respectively along the latitude and longitude directions; and adding the first random noise and the second random noise to the x-coordinate and y-coordinate of each point in the relevant point set, respectively.

[0009] In conjunction with the first aspect, in one possible implementation, the method further includes: grouping the original trajectory dataset according to the start and end positions; determining the similarity between the trajectories in each group; and taking either of the two trajectories with the highest similarity in each group as the initial trajectory to form an initial trajectory dataset.

[0010] In conjunction with the first aspect, in one possible implementation, the probability of directly inferring the calculated point is: ;in, Location visited The number of users, It is the total number of users; the probability of inferring the predicted point when the previous position of the predicted point is known is: ;in, It is a location The number of times it appears, yes The number of times it appears; the probability of inferring the predicted point when the next position after the predicted point is known is: ;in, It is a location The number of times it appears, yes The number of times it appears.

[0011] In conjunction with the first aspect, in one possible implementation, when the relevant point set includes only the sensitive points, only the sensitive region is initialized.

[0012] Secondly, embodiments of this application provide a trajectory privacy protection device based on sensitive regions. The device includes: an iteration module, used to iteratively execute a related point determination step to determine a related point set using preset sensitive points as inference points; a region construction module, used to construct a sensitive region containing the related point set for the preset sensitive points; a connection module, used to sequentially connect the position points within the related point set to form a sensitive sub-trajectory; an addition module, used to add noise to the trajectory points of the sensitive sub-trajectory using a personalized privacy budget allocation method of a differential privacy model; and a generation module, used to replace the portion of the initial trajectory located within the sensitive region with the sensitive sub-trajectory after adding noise, generating a spatiotemporal trajectory dataset to be published. The related point determination step includes: adding the inference point to the related point set; determining whether the probability of inferring the inference point when the preceding or following position point of the inference point is known is greater than the probability of directly inferring the inference point; if the determination result is yes, determining the preceding or following position point of the inference point as the inference point.

[0013] Thirdly, embodiments of this application provide a trajectory privacy protection device based on sensitive areas. The device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the trajectory privacy protection method based on sensitive areas as described in the first aspect or any possible implementation of the first aspect.

[0014] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-readable instructions, which, when executed by a processor, implement the trajectory privacy protection method based on sensitive regions as described in the first aspect or any possible implementation thereof.

[0015] The technical solutions provided in this application embodiment have at least the following technical effects:

[0016] This application provides a trajectory privacy protection method based on sensitive regions. This method protects a set of related points consisting of preset sensitive points and their surrounding points, preventing attackers from launching inference attacks based on background knowledge such as location and spatiotemporal correlation from the noise-protected trajectory, thus preventing the leakage of user privacy. Simultaneously, this method uses a personalized privacy budget allocation method based on a differential privacy model to add noise to the trajectory points of sensitive sub-trajectories. This achieves precise local noise addition while rationally utilizing the privacy budget, improving data availability and reducing the difficulty of data mining. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments of this application or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart illustrating the trajectory privacy protection method based on sensitive regions provided in this application embodiment; Figure 2 A flowchart of the relevant point determination steps provided in the embodiments of this application; Figure 3 A flowchart for constructing sensitive regions provided in this application embodiment; Figure 4 A flowchart for adding noise to trajectory points of sensitive sub-trajectories provided in an embodiment of this application; Figure 5 A flowchart for forming an initial trajectory dataset provided in an embodiment of this application; Figure 6 A schematic diagram of the initialized sensitive region provided in the embodiments of this application; Figure 7 A schematic diagram of the initial trajectory and sensitive areas provided in the embodiments of this application; Figure 8 This is a structural diagram of a trajectory privacy protection device based on sensitive areas provided in an embodiment of this application. Detailed Implementation

[0019] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0020] This application provides a trajectory privacy protection method based on sensitive areas, such as... Figure 1 As shown, the method includes steps S101 to S105.

[0021] S101: Using preset sensitive points as inference points, iteratively execute the relevant point determination steps to determine the relevant point set. Specifically, if key information about a user who has stayed at a certain location can be inferred from that location, then that location is identified as a preset sensitive point; for example, preset sensitive points include schools, hospitals, workplaces, residences, etc. Figure 6 and Figure 7 The example shown illustrates a scenario where the preset sensitive point is a school.

[0022] S102: Construct a sensitive region containing a set of relevant points for the preset sensitive points. For example, Figure 7 The dashed box represents the sensitive area, and the relevant point set within the sensitive area contains 5 location points.

[0023] S103: Connect the location points in the relevant point set sequentially to form a sensitive sub-trajectory. Figure 7 The trajectory of five consecutive location points within the sensitive area is the sensitive sub-trajectory.

[0024] S104: Add noise to the trajectory points of sensitive sub-trajectories using a personalized privacy budget allocation method based on a differential privacy model.

[0025] S105: Replace the portion of the initial trajectory located in the sensitive region with the sensitive sub-trajectory after adding noise, and generate the spatiotemporal trajectory dataset to be published.

[0026] Figure 2 The steps for determining relevant points are shown, including steps S201 to S203.

[0027] S201: Add the calculated points to the relevant point set.

[0028] S202: Determine whether the probability of inferring the calculated point when the previous or next position of the calculated point is known is greater than the probability of directly inferring the calculated point.

[0029] Specifically, the probability of directly inferring the calculated point is: ;in, Location visited The number of users, It is the total number of users; the probability of inferring the calculated point given the previous position of the calculated point is: ;in, It is a location The number of times it appears, yes The number of times it appears; the probability of inferring the predicted point when the next position after the predicted point is known is: ;in, It is a location The number of times it appears, yes The number of times it appears.

[0030] If the judgment result is yes, execute S203: determine the previous or next position point of the calculated point as the calculated point.

[0031] This method protects a set of related points consisting of preset sensitive points and their surrounding points, preventing attackers from launching inference attacks based on background knowledge such as location and spatiotemporal correlation from the noise-protected trajectory, thus preventing the leakage of user privacy. The personalized privacy budget allocation method achieves precise local noise addition and can rationally utilize the privacy budget according to actual conditions, improving data usability and reducing the difficulty of data mining.

[0032] Reference Figure 6 and Figure 7 Sensitive regions containing relevant point sets are constructed based on preset sensitive points, including Figure 3 Steps S301 and S302 are shown.

[0033] S301: Initialize the sensitive area using the latitude and longitude of the preset sensitive point and the adjacent preset distance location points as the boundary using the minimum boundary rectangle. Figure 6 The school is preset as the sensitive point. The sensitive area is initialized using the minimum boundary rectangle, which forms the range shown by the dashed line.

[0034] S302: Modify the sensitive region based on the positions of each point in the relevant point set. The sensitive region needs to cover all points in the relevant point set, and is formed after modification. Figure 7 The area indicated by the dashed line.

[0035] Furthermore, when the relevant point set only includes sensitive points, only the sensitive region is initialized. Figure 6 For example, when the preset sensitive point is a school, and the probability of directly estimating the school is greater than the probability of estimating the school's previous or next location, the sensitive area is... Figure 6 The area indicated by the dashed line.

[0036] The differential privacy model is as follows: ;in, , The number of initial trajectories. For the total privacy budget, A fixed privacy budget is allocated for each sensitive sub-trajectory. For relevant points The allocated privacy budget, A privacy budget scale for any location within a sensitive area. The location entropy; the privacy budget scale for any location within a sensitive area is defined as: ;in, It is the shortest Euclidean distance from the relevant point to the preset sensitive point; the position entropy is: ;in, This represents the proportion of times a user accesses a location out of the total number of times that location is accessed.

[0037] like Figure 4 As shown, adding noise to the trajectory points of the sensitive sub-trajectory includes steps S401 to S403.

[0038] S401: Determine the initial trajectory containing the sensitive sub-trajectory and its position within the initial trajectory.

[0039] S402: Generates a first random noise and a second random noise that follow a Laplace distribution and are distributed along the latitudinal and longitudinal directions, respectively. Specifically, the first random noise, LatitudeNoise, is distributed along the latitudinal direction, and the second random noise, LongitudeNoise, is distributed along the longitudinal direction.

[0040] S403: Add the first random noise and the second random noise to the x-coordinate and y-coordinate of each point in the relevant point set, respectively. Specifically, add the first random noise LatitudeNoise to the x-coordinate of each point in the relevant point set, and add the second random noise LongitudeNoise to the y-coordinate of each point in the relevant point set.

[0041] The method also includes the step of simplifying the original trajectory dataset to obtain an initial trajectory dataset, specifically including... Figure 5 Steps S501 to S503 are shown.

[0042] S501: Group the original trajectory dataset according to the start and end positions.

[0043] Specifically, let the original trajectory dataset be... It consists of m original trajectories, where Let i be the i-th trajectory in the original trajectory dataset D, which contains n trajectory points. for The i-th trajectory point, , , Let be the longitude, latitude, and timestamp of the i-th trajectory point, respectively. The set of starting points of the original trajectory is denoted as . The set of endpoints of the original trajectory is denoted as .

[0044] For each original trajectory in the original trajectory dataset, using the selected original trajectory as the baseline trajectory, the start and end points of the other original trajectories are compared with the baseline trajectory in turn. Original trajectories with the same start and end points are added to the group. In the middle; however, the next baseline trajectory was never added to the group. The original trajectories are selected and compared with the remaining original trajectories in terms of start and end points. This process is repeated until all remaining original trajectories without identical start and end points exist after grouping. These original trajectories are then considered as a group, and the original trajectory dataset is completely grouped. .

[0045] S502: Determine the similarity between trajectories within each group.

[0046] One arbitrary original trajectory in the original trajectory dataset is represented as: Assuming the original trajectory starts from the beginning timestamp up to a certain timestamp The direction angle is ,but ,Right now Another arbitrary original trajectory in the original trajectory dataset is represented as... ,but Therefore, the similarity function between two arbitrary original trajectories is expressed as: , The smaller the value, the more similar the two trajectories are.

[0047] S503: Select either of the two trajectories with the highest similarity within each group as the initial trajectory to form the initial trajectory dataset. Specifically, after determining the original trajectories within each group, iterate through the generated groups to generate an initial trajectory dataset based on similarity metrics, denoted as... .

[0048] While this application provides the method operation steps as described in the embodiments or flowcharts, more or fewer operation steps may be included based on conventional or non-inventive labor. The order of steps listed in this embodiment is merely one possible execution order among many and does not represent the only execution order. In actual device or client product execution, the methods shown in this embodiment or the accompanying drawings can be executed sequentially or in parallel (e.g., in a parallel processor or multi-threaded processing environment).

[0049] This application provides a trajectory privacy protection device based on sensitive areas. The device includes an iteration module 801, a region construction module 802, an addition module 803, and a generation module 804.

[0050] The iteration module 801 is used to iteratively execute the relevant point determination steps to determine the relevant point set, using preset sensitive points as calculation points. The region construction module 802 is used to construct a sensitive region containing the relevant point set for the preset sensitive points. The connection module is used to sequentially connect the position points within the relevant point set to form sensitive sub-trajectories. The addition module 803 is used to add noise to the trajectory points of the sensitive sub-trajectories using a personalized privacy budget allocation method of the differential privacy model. The generation module 804 is used to replace the portion of the initial trajectory located within the sensitive region with the noisy sensitive sub-trajectories to generate a spatiotemporal trajectory dataset to be published.

[0051] The steps for determining relevant points include: adding the inferred point to the relevant point set; determining whether the probability of inferring the inferred point when the previous or next position of the inferred point is known is greater than the probability of directly inferring the inferred point; if the determination result is yes, determining the previous or next position of the inferred point as the inferred point.

[0052] The region construction module 802 is specifically used to: initialize the sensitive region using the latitude and longitude of the preset sensitive point and the adjacent preset distance location points as the boundary with a minimum boundary rectangle; and modify the sensitive region according to the position of each point in the relevant point set.

[0053] The differential privacy model is as follows: ;in, , The number of initial trajectories. For the total privacy budget, A fixed privacy budget is allocated for each sensitive sub-trajectory. For relevant points The allocated privacy budget, A privacy budget scale for any location within a sensitive area. The location entropy; the privacy budget scale for any location within a sensitive area is defined as: ;in, It is the shortest Euclidean distance from the relevant point to the preset sensitive point; the position entropy is: ;in, This represents the proportion of times a user accesses a location out of the total number of times that location is accessed.

[0054] The added module 803 is specifically used to: determine the initial trajectory where the sensitive sub-trajectory is located and its position in the initial trajectory; generate a first random noise and a second random noise that follow a Laplace distribution and are perpendicular to each other; and add the first random noise and the second random noise to the x-coordinate and y-coordinate of each point in the relevant point set, respectively.

[0055] The device also includes a simplification module, which is specifically used to: group the original trajectory dataset according to the start and end positions; determine the similarity between the trajectories in each group; and take any one of the two most similar trajectories in each group as the initial trajectory to form the initial trajectory dataset.

[0056] The probability of directly inferring the calculation point is: ;in, Location visited The number of users, This represents the total number of users; the probability of inferring the predicted point given the previous position of the predicted point is: ;in, It refers to the number of times a position appears. yes The number of times it appears; the probability of inferring the calculated point when the subsequent position of the calculated point is known is: ;in, It is a location The number of times it appears, yes The number of times it appears.

[0057] Furthermore, when the relevant point set only includes sensitive points, only the sensitive region is initialized.

[0058] The apparatus or module described in the above embodiments can be implemented by a computer chip or physical entity, or by a product with a certain function. For ease of description, the above apparatus is described by dividing it into various modules according to their functions. When implementing the embodiments of this application, the functions of each module can be implemented in one or more software and / or hardware. Of course, a module that implements a certain function can also be implemented by combining multiple sub-modules or sub-units.

[0059] Some modules in the apparatus described in this application can be described in the general context of computer-executable instructions that are executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, classes, etc., that perform a specific task or implement a specific abstract data type. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0060] This application also provides a trajectory privacy protection device based on sensitive areas. The device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the trajectory privacy protection method based on sensitive areas as described above in this application.

[0061] This application also provides a computer-readable storage medium storing computer-readable instructions. When the computer-readable instructions are executed by a processor, the trajectory privacy protection method based on sensitive areas described above in this application is provided.

[0062] The aforementioned storage media include, but are not limited to, Random Access Memory (RAM), Read-Only Memory (ROM), Cache, Hard Disk Drive (HDD), or Memory Card. The memory can be used to store computer program instructions.

[0063] The various embodiments described in this specification are presented in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on its differences from other embodiments. All or part of this application can be used in numerous general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, mobile communication terminals, multiprocessor systems, microprocessor-based systems, programmable electronic devices, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices, etc.

[0064] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of this application.

Claims

1. A trajectory privacy protection method based on sensitive areas, characterized in that, include: Using preset sensitive points as calculation points, the relevant point set is determined by iteratively executing the relevant point determination steps; Construct a sensitive region containing the set of relevant points for the preset sensitive points; Connect the location points in the relevant point set sequentially to form a sensitive sub-trajectory; Noise is added to the trajectory points of the sensitive sub-trajectories using a personalized privacy budget allocation method based on a differential privacy model; The portion of the initial trajectory located within the sensitive region is replaced with the sensitive sub-trajectory after noise has been added, generating a spatiotemporal trajectory dataset to be published; The relevant point determination step includes: adding the inferred point to the relevant point set; determining whether the probability of inferring the inferred point when the previous or next position of the inferred point is known is greater than the probability of directly inferring the inferred point; if the determination result is yes, determining the previous or next position of the inferred point as the inferred point; The differential privacy model is as follows: ； in, , The number of the initial trajectories. For the total privacy budget, A fixed-size privacy budget is obtained for each of the aforementioned sensitive sub-trajectories. For relevant points The allocated privacy budget, The privacy budget scale is defined for any location within the sensitive area. For position entropy; The privacy budget metric for any location within the sensitive area is defined as: ;in, It is the shortest Euclidean distance from the relevant point to the preset sensitive point; The position entropy is: ;in, This represents the proportion of times a user accesses a location out of the total number of times a location is accessed. Adding noise to the trajectory points of the sensitive sub-trajectory includes: Determine the initial trajectory in which the sensitive sub-trajectory is located and its position within the initial trajectory; Generate a first random noise and a second random noise that follow a Laplace distribution and are distributed along the latitudinal and longitudinal directions, respectively; The first random noise and the second random noise are respectively added to the x-coordinate and y-coordinate of each point in the relevant point set.

2. The trajectory privacy protection method based on sensitive areas according to claim 1, characterized in that, The construction of a sensitive region containing the relevant point set for the preset sensitive points includes: The sensitive area is initialized using the latitude and longitude of the preset sensitive point and the location points at adjacent preset distances as boundaries, and a minimum boundary rectangle is used. The sensitive region is modified based on the position of each point in the relevant point set.

3. The trajectory privacy protection method based on sensitive areas according to claim 1 further includes: The original trajectory dataset is grouped according to its start and end positions; Determine the similarity between trajectories within each group; Choose one of the two trajectories with the highest similarity in each group as the initial trajectory to form the initial trajectory dataset.

4. The trajectory privacy protection method based on sensitive areas according to claim 1, characterized in that, The probability of directly inferring the calculated point is: ;in, Location visited The number of users, It represents the total number of users; The probability of inferring the predicted point when the previous position of the predicted point is known is: ;in, It is a location The number of times it appears, yes The number of times it appears; The probability of inferring the predicted point when the subsequent position of the predicted point is known is: ;in, It is a location The number of times it appears, yes The number of times it appears.

5. The trajectory privacy protection method based on sensitive areas according to claim 1, characterized in that, When the relevant point set includes only the sensitive points, only the sensitive region is initialized.

6. A trajectory privacy protection device based on sensitive areas, characterized in that, include: The iteration module is used to determine the relevant point set by iteratively executing the relevant point determination steps, with preset sensitive points as the calculation points; A region construction module is used to construct a sensitive region containing the relevant point set for the preset sensitive points; The connection module is used to connect the position points in the relevant point set sequentially to form sensitive sub-trajectories; Add a module to add noise to the trajectory points of the sensitive sub-trajectory using a personalized privacy budget allocation method based on a differential privacy model; The generation module is used to replace the portion of the initial trajectory located within the sensitive region with the sensitive sub-trajectory after adding noise, thereby generating a spatiotemporal trajectory dataset to be published. The relevant point determination step includes: adding the inferred point to the relevant point set; determining whether the probability of inferring the inferred point when the previous or next position of the inferred point is known is greater than the probability of directly inferring the inferred point; if the determination result is yes, determining the previous or next position of the inferred point as the inferred point; The differential privacy model is as follows: ； in, , The number of the initial trajectories. For the total privacy budget, A fixed-size privacy budget is obtained for each of the aforementioned sensitive sub-trajectories. For relevant points The allocated privacy budget, The privacy budget scale is defined for any location within the sensitive area. For position entropy; The privacy budget metric for any location within the sensitive area is defined as: ;in, It is the shortest Euclidean distance from the relevant point to the preset sensitive point; The position entropy is: ;in, This represents the proportion of times a user accesses a location out of the total number of times a location is accessed. Adding noise to the trajectory points of the sensitive sub-trajectory includes: Determine the initial trajectory in which the sensitive sub-trajectory is located and its position within the initial trajectory; Generate a first random noise and a second random noise that follow a Laplace distribution and are distributed along the latitudinal and longitudinal directions, respectively; The first random noise and the second random noise are respectively added to the x-coordinate and y-coordinate of each point in the relevant point set.

7. A trajectory privacy protection device based on sensitive areas, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the trajectory privacy protection method based on any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the trajectory privacy protection method based on sensitive regions as described in any one of claims 1 to 5.

Images